I'd like to take a stab at this. I haven't written technical documentation like this before, and I think it'd be neat to try.
Shoutout to @ax3l, you might be interested - at today's telecon we came up with an idea to allow people to import custom simulation data (not yet sure how, but I'm thinking that they'd have to write a subclass to our abstract Plasma class... unless we figure out something nicer) into PlasmaPy and then having a to_hdf5 method (something I called a "standardizer") that by default saves to OpenPMD. I guess it'd be, in a way, an endorsement for OpenPMD as a standard - which I guess it deserves due to neatness.
From notes:
Make issue ACTUALLY WRITE A PLEP: write a standardizer for data
Decide on standard nomenclature
Especially for (extended) MHD simulations
I'd like to take a stab at this. I haven't written technical documentation like this before, and I think it'd be neat to try.
Shoutout to @ax3l, you might be interested - at today's telecon we came up with an idea to allow people to import custom simulation data (not yet sure how, but I'm thinking that they'd have to write a subclass to our abstract Plasma class... unless we figure out something nicer) into PlasmaPy and then having a
to_hdf5
method (something I called a "standardizer") that by default saves to OpenPMD. I guess it'd be, in a way, an endorsement for OpenPMD as a standard - which I guess it deserves due to neatness.From notes: